StreamingFile limitations
Hi community,
I'm working toward the porting of our code from `BucketingSink<>` to
`StreamingFileSink`.
In this case we use the sink to write AVRO via Parquet and the suggested
implementation of the Sink should be something like:
```
val parquetWriterFactory = Parquet
Hi Enrico,
Sorry for the late reply. I think your understanding is correct.
The best way to do it is to write your own ParquetBulkWriter and the
corresponding factory.
Out of curiosity, I guess that in the BucketingSink you were using the
AvroKeyValueSinkWriter, right?
Cheers,
Kostas
On Fri, Au
Thanks for confirming.
We have a
```
public class ParquetSinkWriter implements Writer
```
that handles the serialization of the data.
We implemented it starting from:
https://medium.com/hadoop-noob/flink-parquet-writer-d127f745b519
https://stackoverflow.com/questions/48098011/how-to-use-apache-fli
